Einige Inhalte dieser Anwendung sind momentan nicht verfügbar.
Wenn diese Situation weiterhin besteht, kontaktieren Sie uns bitte unterFeedback&Kontakt
1. (WO2019042523) METHOD FOR OPERATION OF A RADAR SYSTEM
Anmerkung: Text basiert auf automatischer optischer Zeichenerkennung (OCR). Verwenden Sie bitte aus rechtlichen Gründen die PDF-Version.

Method for operation of a radar system

Patent claims

1 . Method (100) for operating a radar system (1 ) of a vehicle (2), wherein the radar system (1 ) has at least one radar sensor (10) for detecting at least one target (5) outside the vehicle (2), comprising the following steps:

- Performing a prediction of an ego-velocity (vEgo) of the vehicle (2), so that a prediction result is determined,

- Performing a classification for classifying the at least one detected target (5) as a stationary target (5a) using the prediction result, so that a classification result is determined,

- Selecting one of at least two estimation methods (140) for an

estimation of the ego-velocity (vEgo), wherein the selection is dependent on an evaluation of the classification result.

2. Method (100) according to claim 1 ,

characterized in that,

the at least two estimation methods (140) comprise:

- As a first estimation method (140a): Performing a radar based velocity estimation, which is dependent on the at least one target (5) classified as the stationary target (5a), and

- As a second estimation method (140b): Performing an odometry

based velocity estimation, particularly wherein a corrected odometric velocity is used based on a velocity information read from an interface (3) of the vehicle (2).

Method (100) according to claim 1 or 2,

characterized in that,

the evaluation of the classification result on which the selection is dependent comprise:

- Comparing a number of targets (5) classified as being stationary

targets (5a) with a predetermined minimum number of stationary targets (250), wherein a first estimation method (140a) is performed only if this number of targets (5) is higher than or equal to the minimum number of stationary targets (250), and a second estimation method (140b) is performed otherwise, and particularly the second estimation method (140b) is performed only if the predicted ego- velocity (vEgo) is higher than or equal to a predetermined minimum velocity (251 ).

Method (100) according to claim 3,

characterized in that,

the predetermined minimum number of stationary targets (250) is at least 1 or at least 5 or at least 10 or in the range from 1 to 10, preferably 2 to 8, particularly preferably 4 to 6, especially 5.

5. Method (100) according to any of the preceding claims,

characterized in that,

the step of performing the prediction comprises:

- Predicting the ego-velocity (vEgo) and determining corresponding variance information (210) by using Kalman filtering and/or a tracking algorithm.

6. Method (100) according to claim 5,

characterized in that,

the determined variance information (210) is used for performing the classification, particularly by determining a comparison range based on the variance information (210), wherein preferably a relative velocity (vR) of each detected target (5) is compared to the comparison range, and the at least one detected target (5) is classified as a stationary target (5a) if the relative velocity (vR) lies within the comparison range.

7. Method (100) according to any of the preceding claims,

characterized in that,

a first estimation method (140a) comprises:

- Estimating the instantaneous ego-velocity (vEgo) of the vehicle (2) based on the classification, particularly on the targets (5) classified as stationary targets (5a), particularly preferably by using a regression algorithm.

Method (100) according to any of the preceding claims,

characterized in that,

a second estimation method (140b) comprises:

- Estimating the instantaneous ego-velocity (vEgo) of the vehicle (2) based on a corrected odometric velocity, wherein preferably the corrected odometric velocity is determined from an odometric velocity that is corrected by a linear model.

Method (100) according to any of the preceding claims,

characterized in that,

after the step of selecting the selected estimation method (140) is performed, and after the step of performing the selected estimation method (140) at least one parameter of the prediction, particularly for a Kalman-Filtering, is adapted and/or corrected using the estimated ego-velocity (vEgo).

Method (100) according to claim 9,

characterized in that,

before or after the step of adapting the parameter of the prediction, an odometry correction is performed based on a parameter estimation, particularly using a Recursive Least Square approach.

System (1 ) for a vehicle (2), comprising:

- a radar sensor (10) for detecting at least one target (5) outside the vehicle (2),

- at least one control unit (4) for performing the steps of a method (100) according to any of the preceding claims.

Computer program product being stored on a computer readable medium for operating a radar system (1 ), comprising the following:

- Computer readable program means, initiating the computer to perform a prediction of an ego-velocity (vEgo) of the vehicle (2), so that a prediction result is determined,

- Computer readable program means, initiating the computer to perform a classification for classifying the at least one target (5) as a stationary target (5a) using the prediction result, so that a classification result is determined,

- Computer readable program means, initiating the computer to select one of at least two estimation methods (140) for an estimation of the ego-velocity (vEgo), wherein the selection is dependent on an evaluation of the classification result.

Computer program product according to claim 12, wherein a computer readable program means is provided for initiating the computer to perform a method according to any of claims 1 to 10.